Use to measure R performance with profiling and microbenchmarks, design representative workloads, and interpret CPU time, allocations, garbage collection, and scaling. Do not use this skill to choose or implement an optimization without measurement evidence.
